According to the knowledge base, the safe RAID drive replacement/rebuild steps are:
1. Identify which drive failed:
- In SHA: on the active server, go to Storage Manager and check which drive is missing.
- In DSM 7 / storage pool workflows, the relevant article says to use Storage Manager to find the affected storage pool and the failed drive.
2. Replace the failed drive with a healthy one:
- For SHA, install the new drive into the correct slot on the passive server.
- For standard DSM storage pool repair, replace the defective drive with a healthy one.
3. Start the repair:
- Go to Storage Manager > Storage.
- Select the storage pool with Missing or the crashed/degraded pool and click Repair.
4. Confirm the warning:
- The system will warn that it will repair the volume and erase all data on the replaced drive.
5. Wait for synchronization/repair to complete:
- The SHA guide notes that switchover and auto-failover will not be available during synchronization.
Additional notes from the sources:
- If only the read-only cache is missing, data on the passive server should still be available.
- For a crashed storage space or SSD cache, the guide says the drive usually needs to be replaced with a healthy one and the volume recreated.
- The knowledge base also says that, starting from DSM 7.0, a replacement HDD can only replace a defective HDD in a storage pool.
- If Auto Repair is enabled in the relevant system, repair may start automatically after the replacement drive is installed; otherwise, you must click Action > Repair.
Sources used: [1], [15]